On 06/03/15 10:56, Martin Lucina wrote:
Now, there are actually two issues to consider:
1) accurate timekeeping which has >clockintr resolution
2) detecting if interrupts are pending, and yielding the *host
thread* in rump_un/schedule() if so
or, eventually, 3) replace/enhance the -xen and -bmk platforms with a
preemptive host scheduler.
I'd definitely not consider preemptive scheduling an enhancement, which
is why I'm very divided about "2". Let's just admit preemptive
scheduling was a bad idea, perpetuated by the current broken operating
system model, and move towards getting rid of it.
